Infection by hepatitis B virus is a problem of enormous dimensions. Hepatitis B virus has reached epidemic levels worldwide. It is estimated that as many as 350 million people worldwide are persistently infected with HBV, many of whom develop associated pathologies such as chronic hepatic insufficiency, cirrhosis, and hepatocellular carcinoma. After a two to three month incubation period in which the host is unaware of the infection, HBV infection can lead to acute hepatitis and liver damage, that causes abdominal pain, jaundice, and elevated blood levels of certain enzymes. About 1-2% of these develop fulminant hepatitis, a rapidly progressive, often fatal form of the disease in which massive sections of the liver are destroyed, with a mortality rate of 60-70%.
The Epstein-Barr virus is a member of the genus Lymphocryptovirus, which belongs to the subfamily gammaherpesvirine. It is notably lymphotropic. EBV has the classic structure of herpes viruses, viz., its double-stranded DNA genome is contained within an icosapentahedral nucleocapsid, which, in turn, is surrounded by a lipid envelope studded with viral glycoproteins. EBV is now recognized as a cause of B-cell lymnphoproliferative diseases, and has been linked to a variety of other severe and chronic illnesses, including a rare progressive mononucleosis-like syndrome and oral hairy leukoplakia in AIDS patients. The suggestion that EBV is a major cause of chronic fatigue has not withstood scrutiny. EBV is primarily transmitted through saliva, although some infections are transmitted by blood transfusion. More than 85% of patients in the acute phase of infectious mononucleosis secrete EBV.
It has been discovered that certain L-nucleosides, mirror images of the natural DNA constituents may inhibit DNA synthesis at the triphosphate level probably by tight binding to the viral polymerase in the first stage of viral DNA synthesis.
2′-Deoxy-2′-fluoro-β-L-arabinofuranosyl nucleosides have the general formula:
wherein B is a pyrimidine, purine, heterocyclic or heteroaromatic base.Reported Syntheses of L-FMAU
Yung Chi Cheng, Chung K. Chu and others first reported that 1-(2′-deoxy-2′-fluoro-β-L-arabinofuranosyl)-thymine (L-FMAU) exhibits superior activity against hepatitis B virus and Epstein Barr virus in 1994. See U.S. Pat. Nos. 5,587,362; 5,567,688; 5,565,438 and 5,808,040 and International Patent Application published as WO 95/20595.

The Cheng patents describe a synthesis of L-FMAU from the sugar L-xylose (formula A) as well as the sugar L-ribose (formula B).

These patents describe the synthesis of L-FMAU from L-xylose via conversion to the key intermediate 1-O-acetyl-2,3,5-tri-O-benzoyl-β-L-ribofuranose (see for example the '688 patent, starting at column 4, line 62). The key intermediate was synthesized from L-xylose in a total yield of 20% (see also L. Vargha, Chem. Ber., 1954, 87, 1351; Holy, A., et al., Synthetic Procedures in Nucleic Acid Chemistry, V1, 163-67). This synthesis was also reported in Ma, T.; Pai, S. B.; Zhu, Y. L; Lin, T. S.; Shanmunganathan, K.; Du, J. F.; Wang, C. G.; Kim, H.; Newton, G. M.; Cheng, Y. C.; Chu, C. K. J. Med. Chem. 1996, 39, 2835. The inversion of the hydroxy group of L-xylose was achieved via the formation of the 5-O-benzoyl-1,2-O-isopropylidene-α-L-ribofuranoside, followed by a stereoselective hydride transfer during the reduction of the cycloketone furanoside with NaBH4. The resulting ribofuranoside was then converted to 1-O-acetyl-2,3,5-tri-O-benzoyl-β-L-ribofuranose, the key intermediate in the synthesis of L-FMAU (See Scheme A).

1-O-Acetyl-2,3,5-tri-O-benzoyl-β-L-ribofuranose can also be synthesized directly from the more expensive starting material L-ribose (see for example the '688 patent, starting at column 6, line 30; and Holy, A., et al., Synthetic Procedures in Nucleic Acid Chemistry, V1, 163-67). This alternative synthesis of 1-O-acetyl-2,3,5-tri-O-benzoyl-β-L-ribofuranose (yield of 53%) was also reported by Chu, C. K. et al. Antimicrobial Agents Chemother. 1995, 39, 979. This synthetic route to L-FMAU is set out below in Scheme B.

The key intermediate was subsequently fluorinated in a nucleophilic displacement reaction at C2 to obtain 1,3,5-tri-O-benzoyl-2-deoxy-2-fluoro-L-arbinofuranose, which was condensed with a desired base, such as thymine (5-methyluracil) through the bromosugar to provide the 2′-deoxy-2′-fluoro-arabinofuranosyl nucleosides in various yields.
Chu et al. later developed a synthesis for the production of L-FMAU from L-arabinose in 14 steps and an overall yield of 8% (Du, J.; Choi, Y.; Lee, K.; Chun, B. K.; Hong, J. H.; Chu, C. K. Nucleosides and Nucleosides 1999, 18, 187). L-Arabinose was converted to L-ribose in 5 steps (Scheme C). L-Ribose was then used in the synthesis 1-O-acetyl-2,3,5-tri-O-benzoyl-β-L-ribofuranose, which as described above led to the formation of L-FMAU.

The processes mentioned above either start from an expensive sugar (L-ribose or L-xylose) and/or are very long, with low yields. In addition, they involve the use of a nucleophilic form of fluoride such as KHF2 or Et3N-3HF, which is difficult to handle and requires the displacement of an activated hydroxyl group. The instability of DAST prevents its use on large scale. The conversion of 1-O-acetyl-2,3,5-tri-O-benzoyl-β-L-ribofuranose (TBAR) to 1,3,5-tri-O-benzoyl-β-L-ribofuranose generates 2,3,5-tri-O-benzoyl-β-L-ribofuranose as a side-product, though it can be reconverted to TBAR.
Reported Syntheses of 1-O-methyl-2-deoxy-2-fluoro-arabinofuranoside
The synthesis of 1-O-methyl-2-deoxy-2-fluoro-α-D-arabinofuranoside, has been reported by Wright et al. (Wright, J. A.; Taylor, N. F.; Fox, J. J. J. Org. Chem 1969, 34, 2632, and references therein). In this report, D-xylose is used as the starting material, which after a conversion to the corresponding furanose and a series of protection reactions, gave an epoxy furanoside as an intermediate. This compound was further converted to 5-O-benzyl-1-O-methyl-2-deoxy-2-fluoro-α-D-arabinofuranoside, which after removal of the benzyl group afforded 1-O-methyl-2-deoxy-2-fluoro-α-D-arabinofuranoside (Scheme D).

The synthesis of the 1-O-methyl-2-deoxy-2-fluoro-β-D-arabinofuranoside (the anomer of the above compound) was reported by Marquez et al. (Wysocki, R. J.; Siddiqui, M. A.; Barchi, J. J.; Driscoll, J. S.; Marquez, V. E. Synthesis 1991, 1005). D-ribose was converted in several steps to 1,3,5-tri-O-benzoyl-2-deoxy-2-fluoro-β-D-arabinofuranose, the corresponding bromo sugar derivative was produced under HBr/AcOH condition and the reaction of potassium carbonate in methanol gave the desired compound (Scheme E).
Reported Synthesis of 2-deoxy-2-fluoro-D-arabinospyranose
2-deoxy-2-fluoro-D-arabinopyranose was previously made from D-arabinose via D-arabinal as it is shown in Scheme F (Albano, E. L et al. Carbohyd. Res. 1971, 19, 63).

The same material was made from D-Ribose as shown below in Scheme G (Bols, M.; Lundt, I.; Acta Chem. Scand. 1990, 44, 252).
Reported Synthesis of 2-deoxy-2-fluoro-3,4-di-O-acetyl-D-arabinospyranose
The title compound was previously made as a result of an electrophilic addition of selectfluor on D-arabinal (Albert, M. et al, Tetrahedron 1998, 54, 4839; Scheme H).

In light of the commercial importance of L-FMAU, and its use in the treatment of patients afflicted with hepatitis B and Epstein Barr virus, it is an object of the invention to provide an improved synthesis of L-FMAU and related nucleosides.
It is another object of the invention to provide a synthesis of 2′-deoxy-2′-halo-β-L-arabinofuranosyl nucleosides from inexpensive starting materials in relatively high yield.